Structural vibration is one of the key performance limiting phenomena in many types of advanced machinery, such as space launch vehicle shrouds, all types of jet and turbine engines, robots, and many types of manufacturing equipment. For example, semiconductor manufacturing equipment and the equipment used to manufacture micro- and nano-devices are sensitive to structural vibration at ever increasing levels. The positioning accuracy requirements in the most advanced semiconductor manufacturing and test equipment in the market today are on the order of single-digit nanometers.
Because structural vibration depends on many factors that are not easily modeled, such as boundary and continuity conditions, as well as the disturbance environment, it is impossible to design a machine from the first prototype that will meet all vibration requirements. This means that the final steps in analyzing and suppressing vibration are accomplished after the actual production unit has been built. Unfortunately, this is precisely the time that any delay in shipment is the most costly in terms of lost revenue and competitive advantage.
To address this shortfall, it is known to incorporate vibration analysis and suppression systems into precision equipment. In general, a typical vibration analysis and suppression system includes a multitude of vibration sensors and vibration actuators that are installed on-board the precision equipment in selected locations. The system also includes a control system that transmits control signals in accordance with a vibration suppression algorithm to the actuators during normal operation of the precision equipment to mechanically suppress the vibrations. Using a feedback loop, the sensed vibration information is fed back to the control circuitry, which adjusts the control signals in response to dynamic conditions.
Using a combination of feedforward and feedback control theory, the vibration suppression algorithm used by the control circuitry to generate the control signals is selected in accordance with vibration information acquired by the sensors during vibration testing of the precision equipment, preferably before the precision equipment is operated in the field.
The vibration analysis portion of this process is typically implemented during initial vibration testing. In particular, sensors are affixed to select locations on the precision equipment, and operated to sense the response of the precision equipment to artificially induced environmental vibrations. The sensed vibrations are then analyzed to ascertain the nature of the vibration suppression algorithm to be programmed into the control circuitry. Once the algorithm has been programmed, at least some of the sensors will then be replaced with the actuators that will be used to generate the control signals that suppress the environmental vibrations during the feedback control portion of the vibration testing process (to ascertain performance of the control algorithm), as well as during normal operation of the precision equipment in the field (to improve performance by suppressing vibrations at key locations).
The feedback and feedforward control portions of the vibration suppression process are typically implemented both during vibration testing after the vibration suppression algorithm has been programmed into the control circuitry, as well as during normal operation of the precision equipment in the field. In particular, in response to the normal operating environment, the sensors feed back vibration information to the control circuitry, which in response, generates the vibration suppression control signals, the parameters of which are continually adjusted in real-time in response to the varying vibration conditions. In case of predictable and repeated disturbance, such as from cooling fans, etc., the disturbance information as measured by dedicated sensors can be fed forward to the controller to improve control performance even further. These control signals are transmitted to the actuators, which vibrate to suppress the environmental vibrations. Minor adjustments of the vibration suppression algorithm as previously designed can then be performed based on the actual performance of the vibration suppression system.
There are several unresolved issues that can be addressed during vibration testing. For example, because the sensors must be affixed to key locations in the precision equipment in a robust manner (typically using a bonding material, such as epoxy) to ensure the accuracy of the sensed vibration information during feedforward vibration testing, replacement of the sensors with actuators for feedback vibration testing and normal operation of the precision equipment can be a tedious process. In addition, the analysis of the sensed vibration information and programming of the vibration suppression algorithm may sometimes be accomplished by third parties that are remote from the equipment site, and who must, therefore, repeatedly interface with personnel on-site during the iterative vibration information acquisition and algorithm programming process.
There are also unresolved issues that can be addressed during normal operation of the precision equipment in the field. For example, control circuitry currently used in vibration analysis and suppression systems is located remotely from the vibrating part of the precision equipment, typically being hardwired to the on-board sensors and actuators even during the normal operation of the precision equipment. There are several disadvantages to this architecture.
For example, the connecting cables extending from the precision equipment stationary parts, such as electronics cabinets, to the moving parts, such as stages and end effectors, often hinder its normal operation-especially in the case where the sensors and actuators are located on rotating or rapidly translating components. Such cables may in fact introduce unwanted vibrations, at least partially negating the benefits of the vibration suppression system. Significantly, because most vibration actuators, such as piezoceramics, voice coils and others, require a relatively large voltage (typically in the hundreds of volts), the cables are quite bulky, providing a further hindrance to normal operation of the equipment. In addition to the mechanical awkwardness, the use of cables (both from the sensors and to the actuators) also provides a long path through an environment rich in electromagnetic noise that can be injected into the analog signals transmitted between the sensors/actuators and the remotely located control circuitry. As a result, proper control of the actuators may be compromised due to corruption of either the control signals transmitted from the control circuitry to the actuators or the sensing signals transmitted to the control circuitry from the sensors, or both.
There thus remains a need for improved vibration analysis and suppression systems and methods for testing and implementing such systems.